基于 MaxEnt 模型的新疆松材线虫病入侵风险预测与评估

Risk analysis/assessment of pine wilt disease invasion in Xinjiang based on MaxEnt model prediction

  • 摘要:
    目的 松材线虫病由松材线虫引发,以新疆本地墨天牛属昆虫为主要传播媒介,可对天然松林毁灭性破坏。本研究以新疆为研究区域,聚焦松材线虫病及其媒介昆虫适生分布特征,开展入侵风险预测与综合评估,为新疆松材线虫病监测预警、源头防控和区域联防联控提供参考依据。
    方法 本文基于松林内分布有云杉小墨天牛(Monochamus sutor)和云杉大墨天牛(Monochamus sartor urussovi)的地理分布点和环境因子,采用最大熵模型(MaxEnt)和ArcGIS对新疆地区松材线虫病潜在适生区进行预测,并使用刀切法对影响松材线虫病潜在适生区分布的关键环境变量进行筛选。运用多指标综合评估方法,对松材线虫病的入侵与危害风险性进行综合分析评估。
    结果 MaxEnt模型中的AUC>0.9,表明模型预测结果具有较高可靠性,可用于松材线虫病潜在侵入风险预测。Bio1年平均气温、Bio3等温性、Bio11最冷季度平均温度、Bio10最暖季度平均温度、Bio5最暖月最高温度、Bio9最干季度平均温度、Bio6最冷月最低温度是影响松材线虫病潜在发生的主要环境变量。该模型预测松材线虫病潜在发生的高适生区主要集中于阿尔泰山、天山、昆仑山三大山系。高、中、低适生区面积分别占新疆总面积的 18.35%、11.12%、8.57%,总计 38.22%。应用多指标综合评估方法对其扩散蔓延的风险评估结果表明,松材线虫病是新疆重要的危险性森林有害生物,其风险值R为2.64,属特别危险的森林有害生物。
    结论 松材线虫病在新疆的潜在适宜分布区域广泛,适生区面积占新疆总面积的38.22%,继续扩张和大面积暴发的潜在风险较高。研究可为相关部门合理制定预防措施提供理论依据。

     

    Abstract:
    Objective Pine wilt disease (PWD) is caused by Bursaphelenchus xylophilus and mainly transmitted by insects of the genus Monochamus, which can cause devastating damage to pine forests. Taking Xinjiang as the research area, this study focused on the suitable distribution characteristics of pine wilt disease and its vector insects, carried out invasion risk prediction and comprehensive evaluation, so as to provide scientific reference for the monitoring, early warning, source prevention and control, and regional joint prevention and control of pine wilt disease in Xinjiang.
    Methods This paper is based on the geographical distribution points and environmental factors of Monochamus sutor and Monochamus sartor urussovi in pine forests. this paper uses the MaxEnt and ArcGIS to predict the distribution of potentially suitable areas for PWD in Xinjiang and uses the knife-cut method to filter the key environmental variables affecting the distribution of potentially suitable areas for PWD. The key environmental variables affecting the distribution of potential pine wood nematode habitat were screened using the knife-cut method. The risk of invasion and damage of PWD was analysed and evaluated using a comprehensive multi-indicator assessment method.
    Results The AUC in the MaxEnt model was 0.998, indicating that the model predictions are highly reliable and can be used to predict the potential invasion risk of PWD. Bio1 mean annual temperature, Bio3 isothermality, Bio11 coldest quarterly mean temperature, Bio10 warmest quarterly mean temperature, Bio5 warmest monthly maximum temperature, Bio9 driest quarterly mean temperature, and Bio6 coldest monthly minimum temperature were the main environmental variables influencing the potential incidence of pine nematode disease. The model predicted that the high aptitude zones for the potential occurrence of PWD were mainly concentrated in the three mountain systems of Altai Mountains, Tianshan Mountains and Kunlun Mountains. Its high, medium and low suitability areas accounted for 18.35%、11.12%、8.57% of the total area of Xinjiang, respectively. The results of the risk assessment of the spread of the disease by applying a comprehensive multi-indicator assessment method show that PWD is an important and dangerous forest pest in Xinjiang, with a risk value of 2.64, making it a particularly dangerous forest pest.
    Conclusion PWD has a wide range of potentially suitable distribution areas in Xinjiang, with the area of suitable areas accounting for 38.22% of the total area of Xinjiang, and the potential risk of continued expansion and large-scale outbreaks is high. The study can provide a theoretical basis for the relevant departments to rationally formulate preventive measures.

     

/

返回文章
返回